Approximate Dynamic Programming for Optimizing Oil Production

In this chapter, a new ADP algorithm integrating (1) systematic basis function construction, (2) a linear programming (LP) approach in DP, (3) adaptive basis function selection and (4) bootstrapping, is developed and applied to oil production problems. The procedure requires the solution of a large-scale dynamic system, which is accomplished using a subsurface flow simulator, for function evaluations. Optimization results are presented for cases involving single-phase primary oil production and water injection. In the first case the global optimum can be computed, and the ADP results are shown to essentially achieve this optimum. Clear improvement, relative to various baseline strategies, is similarly observed in the second case. Components D R A F T February 17, 2012, 1:35pm D R A F T 2 APPROXIMATE DYNAMIC PROGRAMMING FOR OPTIMIZING OIL PRODUCTION of the new algorithm, or even the entire procedure, may also be applicable in other domains.